AMD Radeon Pro WX 9100 Review | 118 Data compared

The AMD Radeon Pro WX 8200 is a highly competitive sub-£1,000 workstation graphics card built on the Vega 10 architecture. It features 3,584 stream processors, a 1,200 MHz base clock, and 8GB of innovative HBM2 memory on a wide 2,048-bit bus, delivering an impressive 512 GB/s of memory bandwidth. In testing, the card excels remarkably in GPU Compute and OpenCL-heavy tasks, outperforming its main rival, the NVIDIA Quadro P4000, and even beating the more expensive Quadro P5000 in benchmarks like LuxMark and Blender. Furthermore, it demonstrates an exceptional ability to handle heavy multi-tasking, allowing users to continue smooth 3D modelling in applications like Autodesk 3ds Max or Maya while simultaneously running intensive background video encodes. However, the card does come with a few notable limitations. Its primary drawback is its high power consumption, carrying a 230W TDP rating that demands both 8-pin and 6-pin power connections, making it much less power-efficient than the NVIDIA competition. While the 8GB frame buffer is fast, it can cause the card to fail or crash when rendering highly intensive production scenes that exceed this...

Here is a detailed summary of the 2nd Generation AMD Ryzen Threadripper review: Summary of the 2nd Gen AMD Ryzen Threadripper Review The 2nd Generation AMD Ryzen Threadripper series, built on the 12nm Zen+ architecture, represents a massive paradigm shift for professional design visualisation by making ultra-high core counts affordable on a single desktop CPU. Historically, rendering tasks required prohibitively expensive £10k dual Intel Xeon configurations. However, AMD's new line up disrupts this space by splitting processors into the X-series (up to 16 cores) for enthusiasts and the powerhouse WX-series (up to 32 cores) targeted at creators. The flagship 32-core Threadripper 2990WX acts as the standout performer in the review, combining a 3.0GHz base frequency with a 4.2GHz turbo boost to match or outpace competitive options from Intel at a much lower entry point of $1,799. The hardware exhibits a clear balance of distinct performance trade-offs across professional workloads. On the positive side (pros), the Threadripper 2990WX delivers phenomenal rendering speeds, tearing through a Luxion KeyShot 4K test scene in just 116 seconds—outperforming an overclocked 18-core Intel Core i9-7980XE by...

This article reviews the second-generation AMD Ryzen Threadripper workstation processors, focusing on the flagship 32-core Threadripper 2990WX built on the 12nm Zen+ architecture. Designed specifically for creators and design visualization professionals, the CPU delivers phenomenal multi-threaded rendering performance that rivals expensive dual-Intel Xeon server setups at a fraction of the cost. In benchmarks like Luxion KeyShot and V-Ray, all 64 threads easily chew through heavy ray-tracing tasks, offering designers near-instant feedback. Crucially, AMD manages to sustain high performance without severely penalizing single-threaded workflows. The processor maintains a 4.2GHz turbo frequency during single-threaded CAD tasks in SolidWorks, ensuring that 3D viewport manipulation remains smooth even with large models. However, the 2990WX comes with distinct trade-offs, primarily its massive 250W Thermal Design Power (TDP). This high power draw demands robust liquid cooling systems and high-wattage power supplies, especially since sustained heavy rendering causes frequencies to drop from an initial 3.30GHz down to 2.90GHz as the processor heats up. While the chip boasts aggressive automatic boosting technologies and extensive manual overclocking overhead, doing so invalidates AMD's warranty and risks system instability. Additionally, for...

The AMD Radeon Pro WX 9100 eGPU workstation card, housed inside a Sonnet eGFX 650 Breakaway Box, provides a substantial performance boost when paired with the 2018 15-inch i9 MacBook Pro. In synthetic benchmarks and timeline intensive workflows, the eGPU excels, scoring over double the internal Radeon Pro 560X graphics card in Geekbench 4’s OpenCL test and Unigine’s Heaven gaming benchmark. The pros of this setup are most evident when handling heavy, bottlenecking codecs. For example, rendering Canon Cinema RAW Lite in DaVinci Resolve dropped from 20 minutes to just 8 minutes, and temporal noise reduction times were cut down by more than half. Furthermore, tasks requiring 3D rendering in Blender saw massive improvements, with the render completion time plummeting from nearly 27 minutes on the MacBook alone down to just 5 minutes. Despite these dramatic speeds, the configuration comes with notable cons regarding cost and specific workflow efficiency. At a steep price point of $1,500 for the graphics card alone—plus an additional $399 for the Sonnet enclosure—it is a heavy investment that may not benefit standard consumers. Testing revealed...

Stable Diffusion Performance (NVIDIA RTX vs. Radeon PRO) This 2023 Puget Systems review evaluates the local performance of professional-grade AMD Radeon PRO and NVIDIA RTX GPUs across three Stable Diffusion implementations: Automatic 1111, SHARK, and PugetBench. The testing platform utilized an AMD Threadripper PRO 5975WX CPU. Results reveal that software optimization dictates hardware success. In the widely used Automatic 1111 web UI (with xFormers), NVIDIA dramatically outperforms AMD. The slowest NVIDIA card tested (RTX A5000) achieves over three times the iterations per second of the fastest AMD card (Radeon PRO VII). However, the ecosystem flips completely in the SHARK framework, where the AMD Radeon PRO W7900 secures the top spot with 18.53 iterations per second, beating out the flagship NVIDIA RTX 6000 Ada Generation. The review highlights distinct pros and cons for both GPU manufacturers based on workflow requirements. NVIDIA's primary advantages include widespread developer priority, commanding raw speed on the top-tier RTX 6000 Ada (21.11 it/s in Automatic 1111), and vastly superior budget efficiency for pure image generation via the 24GB RTX A5000. The primary con for NVIDIA is...

The AMD Radeon Pro WX 8200 is a double-slot high-end professional workstation GPU built on the Vega architecture, offering a lower-cost alternative to the flagship WX 9100. Equipped with 3,584 Stream processors and 8GB of fast HBM2 ECC memory, it targets demanding workloads like real-time visualization, virtual reality (VR), and photorealistic rendering. Its standout feature is superior multitasking enabled by AMD’s Asynchronous Compute Engine, allowing users to run heavy ray-trace rendering tasks at 100% load while maintaining a fluid, lag-free 3D viewport in CAD applications. Additionally, its High-Bandwidth Cache Controller (HBCC) acts as a distinct pro by allowing memory-heavy scenes to safely spill over into system memory, preventing application crashes when exceeding the physical 8GB limit. Despite these workflow innovations, the GPU suffers from notable cons regarding power and pure performance metrics. It has a high Thermal Design Power (TDP) of 230W, requiring a beefy power supply with both 6-pin and 8-pin external connectors, which limits its compatibility to mid-to-high-end desktop towers. Performance testing shows a mixed bag: while it excels and beats the more expensive Nvidia Quadro P5000 in...

The AMD Radeon Vega GPU architecture, evaluated through the Radeon Pro WX 9100 and the Radeon Vega Frontier Edition, delivers highly capable performance across 3D CAD, VR, and GPU rendering workflows. Testing in SolidWorks 2015 and PTC Creo 3.0 shows that both cards provide robust power for engineering software, with the Frontier Edition successfully recognized for advanced features like RealView. Additional strengths include strong performance in DirectX 12 applications, excellent VR-ready capabilities with an integrated HDMI port, and a competitive edge in Radeon ProRender tasks due to optimized OpenCL performance. The Frontier Edition, in particular, offers impressive raw performance and high clock speeds at a highly competitive price point of £758 + VAT. Despite these strengths, the Vega lineup presents distinct drawbacks depending on the specific model chosen. Both cards are power-hungry, demanding up to 300W for the Frontier Edition and 230W for the WX 9100, meaning users must possess substantial workstation power supplies and enough physical chassis space for full-length, double-height cards. The budget-friendly Frontier Edition lacks critical independent software vendor (ISV) certifications and workstation OEM support, and features...